Vehicle occupant classification using radar point cloud

Abstract

A system comprises a computer having a processor and a memory, the memory storing instructions executable by the processor to access sensor data from one or more imaging radar sensors of a vehicle, identify, based on the sensor data, a radar point cloud corresponding to an occupant of the vehicle, analyze the radar point cloud to determine an occupant classification of the occupant, determine an operating parameter for a feature of the vehicle based on the occupant classification, and implement the operating parameter for the feature.

What is claimed is: 
     
       1. A system, comprising:
 a computer having a processor and a memory, the memory storing instructions executable by the processor to: 
 access sensor data from one or more imaging radar sensors of a vehicle; 
 identify, based on the sensor data, a radar point cloud corresponding to an occupant of the vehicle; 
 analyze the radar point cloud to determine an occupant classification of the occupant, including determining coordinates of a center of the radar point cloud according to a reference coordinate system, and determining the occupant classification of the occupant based on the coordinates of the center of the radar point cloud; 
 determine to disable a passive restraint system in the vehicle based on the occupant classification determined by analyzing the radar point cloud; and 
 disable the passive restraint system. 
 
     
     
       2. The system of  claim 1 , the memory storing instructions executable by the processor to:
 determine a height of the radar point cloud; and 
 determine the occupant classification of the occupant further based on the height of the radar point cloud. 
 
     
     
       3. The system of  claim 1 , the memory storing instructions executable by the processor to:
 determine measurements of the radar point cloud in multiple dimensions; and 
 determine the occupant classification of the occupant further based on the measurements of the radar point cloud in the multiple dimensions. 
 
     
     
       4. The system of  claim 1 , the memory storing instructions executable by the processor to determine the occupant classification of the occupant based on the coordinates of the center of the radar point cloud according to the reference coordinate system and one or more of:
 a height of the radar point cloud; 
 measurements of the radar point cloud in multiple dimensions; or 
 a number of points in the radar point cloud. 
 
     
     
       5. The system of  claim 1 , the determining the occupant classification of the occupant including identifying a class of the occupant from among a plurality of classes. 
     
     
       6. The system of  claim 5 , the plurality of classes including multiple classes corresponding to different respective percentile ranges of human size. 
     
     
       7. The system of  claim 5 , the plurality of classes including at least one child class. 
     
     
       8. The system of  claim 5 , the plurality of classes including at least one pet class. 
     
     
       9. A method, comprising:
 accessing sensor data from one or more imaging radar sensors of a vehicle; 
 identifying, based on the sensor data, a radar point cloud corresponding to an occupant of the vehicle; 
 analyzing the radar point cloud to determine an occupant classification of the occupant, including determining coordinates of a center of the radar point cloud according to a reference coordinate system, and determining the occupant classification of the occupant based on the coordinates of the center of the radar point cloud; 
 determining to disable a passive restraint system in the vehicle based on the occupant classification determined by analyzing the radar point cloud; and 
 disable the passive restraint system. 
 
     
     
       10. The method of  claim 9 , comprising:
 determining a height of the radar point cloud; and 
 determining the occupant classification of the occupant further based on the height of the radar point cloud. 
 
     
     
       11. The method of  claim 9 , comprising:
 determining measurements of the radar point cloud in multiple dimensions; and 
 determining the occupant classification of the occupant further based on the measurements of the radar point cloud in the multiple dimensions. 
 
     
     
       12. The method of  claim 9 , comprising determining the occupant classification of the occupant based on the coordinates of the center of the radar point cloud according to the reference coordinate system and one or more of:
 a height of the radar point cloud; 
 measurements of the radar point cloud in multiple dimensions; 
 a number of points in the radar point cloud; or 
 coordinates of a center of the radar point cloud according to a reference coordinate system. 
 
     
     
       13. The method of  claim 9 , the determining the occupant classification of the occupant including identifying a class of the occupant from among a plurality of classes. 
     
     
       14. The method of  claim 13 , the plurality of classes including multiple classes corresponding to different respective percentile ranges of human size. 
     
     
       15. The method of  claim 13 , the plurality of classes including at least one child class. 
     
     
       16. The method of  claim 13 , the plurality of classes including at least one pet class. 
     
     
       17. A system, comprising:
 a computer having a processor and a memory, the memory storing instructions executable by the processor to: 
 access sensor data from one or more imaging radar sensors of a vehicle; 
 identify, based on the sensor data, a radar point cloud corresponding to an occupant of the vehicle; 
 analyze the radar point cloud to determine an occupant classification of the occupant, including determining a number of points in the radar point cloud, and determining the occupant classification of the occupant based on the number of points in the radar point cloud; 
 determine to disable a passive restraint system in the vehicle based on the occupant classification determined by analyzing the radar point cloud; and 
 disable the passive restraint system. 
 
     
     
       18. The system of  claim 17 , the memory storing instructions executable by the processor to determine the occupant classification of the occupant based on the coordinates of the center of the radar point cloud according to the reference coordinate system and one or more of:
 a height of the radar point cloud; 
 measurements of the radar point cloud in multiple dimensions; or 
 a number of points in the radar point cloud. 
 
     
     
       19. The system of  claim 17 , the determining the occupant classification of the occupant including identifying a class of the occupant from among a plurality of classes. 
     
     
       20. The system of  claim 19 , the plurality of classes including at least one of:
 multiple classes corresponding to different respective percentile ranges of human size, 
 at least one child class; or 
 at least one pet class.